Spritecook Generate Tilesets logo

Spritecook Generate Tilesets

Organization
SpriteCook
spritecook-generate-tilesets

Tileset generation guide for SpriteCook. Use with spritecook-workflow-essentials when generating autotile tilesets through SpriteCook MCP tools, choosing tile sizes, using reference/edit/style asset IDs, and saving generated tileset asset IDs.

Overview

PublisherSpriteCook
Repositoryskills
Skill namespritecook-generate-tilesets
Stars
51
Forks
4
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by SpriteCook on GitHub. Read the source before you install it.

Installation

Install the Spritecook Generate Tilesets AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/SpriteCook/skills/tree/main/skills/spritecook-generate-tilesets
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/SpriteCook/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/spritecook-generate-tilesets .claude/skills/spritecook-generate-tilesets
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Spritecook Generate Tilesets in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Spritecook Generate Tilesets on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Spritecook Generate Tilesets is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

SpriteCook Generate Tilesets

Use this skill for SpriteCook tileset generation. Pair it with spritecook-workflow-essentials for credits, manifests, safe downloads, and asset tracking.

Requires: A connected SpriteCook MCP server. If disconnected, explain that connection is needed and use the host's native connection flow when authorized. Do not install software or change authentication settings as part of an art task.

Tools

list_tileset_options

Call this when you need current supported perspectives, piece sets, tile sizes, elevations, edge modes, output dimensions, or defaults.

generate_tileset

Generate a game-ready tileset. The tool returns a job immediately by default; follow the returned poll.tool and poll.arguments until canonical assets entries are ready.

ParameterTypeDefaultDescription
promptstringrequiredTerrain/material request, e.g. mossy dungeon floor
style_modestringpixelpixel or detailed
perspectivestringtopdowntopdown, platformer, or isometric
piece_setstringregistry default15-piece, 17-piece-base, autotile-16-set, isometric-3x5-32, or isometric-2x4-64
tile_sizeintregistry defaultFinal tile size in pixels
elevationstringregistry defaultno-elevation or minimal where supported
edgesstringtransparenttransparent or two_surfaces; two_surfaces only works for 15-piece top-down
variationsint1Number of variations, 1-4
modelstringnullOptional model override. Omit to use SpriteCook's tileset default
colorsstring[]nullOptional hex color guidance, max 64
force_enabledboolfalseForce the output toward force_colors
force_colorsstring[]nullOptional forced hex palette, max 64
project_idstringnullOwned project to place the tileset in
reference_asset_idstringnullExisting tileset asset to use as source/reference; tileset settings are inherited
edit_asset_idstringnullExisting tileset asset to edit; tileset settings are inherited
style_asset_idstringnullExisting asset to use as visual style guide only
wait_secondsint0Optional bounded wait from 0-90 seconds before returning the polling contract

reference_asset_id and edit_asset_id are mutually exclusive. The referenced asset must belong to the SpriteCook account.

Recommended Defaults

  • For top-down pixel autotiles, start with style_mode="pixel", perspective="topdown", piece_set="15-piece", tile_size=32, edges="transparent".
  • For top-down inner-corner base tiles, use piece_set="17-piece-base" and keep edges="transparent".
  • For side-view platformers, use perspective="platformer", piece_set="autotile-16-set", edges="transparent".
  • For detailed top-down tilesets, use style_mode="detailed" and call list_tileset_options before choosing size/elevation.
  • Use model only when the user explicitly wants to compare models.

Reference Workflow

  • When the user asks to use a saved tileset preset, use list_presets(mode="tileset", query=...) and get_preset_settings first, then map the returned tileset settings into generate_tileset.
  • If the user has a local image file path, use spritecook-upload-assets first, then pass the returned asset ID as reference_asset_id, edit_asset_id, or style_asset_id.
  • If the user supplies a small data URL or raw base64 value, call import_asset first, then pass the returned asset ID as reference_asset_id, edit_asset_id, or style_asset_id.
  • Use reference_asset_id when the existing tileset should guide a new generation while preserving its tile size/layout.
  • Use edit_asset_id when the user wants a direct change to an existing tileset.
  • Use style_asset_id as a style guide image when the image should affect only visual style, palette, proportions, and rendering, not tileset layout. The prompt does not need to repeat that the image is a style guide unless the user asks to emphasize a specific detail.
  • When referencing or editing a tileset, do not change style_mode, perspective, piece_set, tile_size, or elevation; SpriteCook inherits and locks those settings.
  • edges is not inherited. When the source is a two_surfaces set, pass edges="two_surfaces" again on every edit or reference generation; otherwise the request falls back to transparent.
  • Edits and reference generations skip the layout guide grid that fresh generations use, so they fail or produce pieces that don't tile more often. An edit can fail with tileset_unusable; the credits are refunded. When an edit fails twice, generate from the set as reference_asset_id or start fresh with a clearer prompt instead of retrying the same edit.

Prompting

  • Keep prompts short and material-focused: snowy stone path, muddy swamp grass, volcanic rock, clean wooden floor.
  • For single-surface tilesets, name one main material.
  • For two_surfaces, name both surfaces clearly: grass and water, volcanic rock and lava.
  • Avoid asking for labels, UI, characters, props, or scene composition inside the tileset.

Output Handling

  • Pass wait_seconds=90 on generate_tileset; most tilesets finish inside it, so polling is rarely needed. Otherwise follow the returned polling contract with check_job_status until the job reaches a terminal state.
  • Check every variation before using it. About one in three can have a full interior tile that doesn't wrap (opposite edges differ in brightness) or pieces that don't meet cleanly, and the response doesn't flag it. Compare the full tile's left and right edges and its top and bottom edges, and paint a small test map with spritecook-use-dual-grid-tilesets for 15-piece sets. A full tile that doesn't wrap shows as a checkerboard over large areas. Pick a clean variation or regenerate.
  • 15-piece sets keep the grid's stepped silhouette on diagonals. Shorelines get a soft edge, but a diagonal coast still reads as steps.
  • Save each returned asset_id in the project manifest or task notes.
  • Use sprite_url as the canonical downloadable tileset image.
  • Treat url, pixel_url, and raw_url as compatibility aliases.
  • If a successful response contains warning.code="asset_output_unavailable", execute the supplied warning.recovery tool call.

Frequently asked questions

What does the Spritecook Generate Tilesets AI skill do?

Tileset generation guide for SpriteCook. Use with spritecook-workflow-essentials when generating autotile tilesets through SpriteCook MCP tools, choosing tile sizes, using reference/edit/style asset IDs, and saving generated tileset asset IDs.

Why use Spritecook Generate Tilesets on TypingMind?

Because you install it once and use it with any model. Spritecook Generate Tilesets is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Spritecook Generate Tilesets in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SpriteCook/skills/tree/main/skills/spritecook-generate-tilesets. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Spritecook Generate Tilesets?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Spritecook Generate Tilesets?

As many as you like. As long as a model supports skills, you can use Spritecook Generate Tilesets with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Spritecook Generate Tilesets AI skill free?

Yes. It is published on GitHub by SpriteCook under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇